Abstract
Miracle fruit (Synsepalum dulcificum) is the botanical source of miraculin, a natural, noncaloric sweetener. Miracle fruit plants have a bush-like architecture and produce multiple flushes of attractive red berries each year. The berries consist of a large seed, opaque pulp, and brilliant red peel. The pulp of the fruit contains a glycoprotein, miraculin, that binds to the tongue’s sweet receptors and induces a conformational change in response to acidic stimuli. Thus, a strong sweet sensation is imparted in the absence of sugars. The miracle fruit plant is becoming increasingly popular because of its taste-modifying properties, but the species lacks many of the breeding tools common to other crops. We report miracle fruit pulp transcriptomes from ‘Sangria’, ‘Vermilion’, ‘Flame’, and ‘Cherry’ morphotypes. A consensus transcriptome included 91,856 transcripts. Reads mapping to the miraculin gene had the highest representation in individual miracle fruit pulp transcriptomes. Other abundant transcripts primarily included Gene Ontology categories representing cellular components, nucleus and nucleic acid binding, and protein modification. The transcriptomes were used to design real-time quantitative reverse-transcription polymerase chain reaction (qRT-PCR) primers for actin, elongation factor 1α, and the miraculin gene. Analysis by qRT-PCR indicated that miracle fruit pulp and peel tissues had the highest abundance of miraculin transcripts, although other tissues such as leaf, root, and flower also had detectable levels of the target sequence. Overall, these results will support discovery research for miracle fruit and the eventual breeding of this species.
Miracle fruit (Synsepalum dulcificum) is native to West Africa, where it has been used for more than 100 years to increase the palatability of local foods. The center of diversity includes regions around Nigeria, Ghana, and Benin. The active ingredient in miracle fruit, miraculin, is a glycoprotein that changes the sensory perception of sour (acidic) to sweet. Although relatively tasteless itself, miraculin has a sweetening effect on acidic food and beverages that is estimated to be 400,000-times greater than that of table sugar (Kurihara and Beidler 1968; Rodrigues et al. 2016). The effect lasts from a minimum of 30 min to 1 h (Inglett et al. 1965; Kurihara and Beidler 1968; Rodrigues et al. 2016; Theerasilp et al. 1989). This natural, noncaloric sweetener shows great promise for reducing sugar in foods and beverages (Achigan-Dako et al. 2015; Gibbs et al. 1996; Kant 2005; Temussi 2006). Fresh fruit and dehydrated miracle fruit pulp tablets are available through online vendors for consumers seeking novel sensory experiences. Additionally, miracle fruit tablets are being used to reduce the negative sensory side effects experienced by chemotherapy patients, thus increasing caloric intake (Wilken and Satiroff 2012). Although the sensory impacts of miraculin are well-studied, foundational genomics studies could greatly advance our understanding of this intriguing species.
Obesity is one of the greatest health challenges facing developed nations (Malik et al. 2013), and miracle fruit could be a part of a holistic solution. Worldwide, 2.1 billion people are obese or overweight, including 35% of adults and 33% of children in the United States (Smith and Smith 2016). The health impacts of obesity include millions of premature deaths annually. Natural, noncaloric sweeteners may be one way to reduce sugar consumption. In general, compared with artificial sweeteners, natural sweeteners are growing in popularity. Steviol glycosides from stevia (Stevia rebaudiana) leaves are now found in many foods and beverages. These glycosides are ∼200- to 400-times sweeter than table sugar and are heat-stable; however, they can also be very bitter. Mogrosides from monk fruit (Siraitia grosvenorii) are also natural, noncaloric sweeteners that are 100- to 250-times sweeter than table sugar, although access to diverse material is limited outside of China. Conversely, miracle fruit has no bitter sensory impact. Consumer demand for miracle fruit could increase as markets shift toward the inclusion of more botanically sourced ingredients.
Historically, miracle fruit, like many other tropical species, has lacked foundational molecular, genetic, and genomics information to fully leverage the diversity of this species and create novel, more profitable cultivars (Achigan-Dako et al. 2015; Xingwei et al. 2016). Recently, however, excellent resources have been published for this species. The chloroplast genome of 1.6 Mb with a total of 133 genes including 88 protein-coding genes is publicly available (Niu et al. 2020). A miracle fruit chromosome-level genome assembly of ∼550 Mb containing 37,911 genes along with transcriptomes of various miracle fruit tissues is also available (Yang et al. 2022). Nevertheless, most molecular miracle fruit studies have focused on the miraculin protein rather than genetics or diversity analyses.
Although true genetic diversity for miracle fruit is probably lacking outside the center of diversity, plants with morphological diversity have been characterized for yield and miraculin content (Demesyeux et al. 2020). This variability could potentially be used for the genetic interrogation of important agronomic and quality traits. The purpose of this study was to generate miracle fruit pulp transcriptomes from morphologically distinct miracle fruit accessions to advance the genomics data available for this species. The results of this study have the potential to aid the genetic improvement of miracle fruit to support growing interest in miraculin as a natural, noncaloric sweetener.
Materials and Methods
Plant materials.
Miracle fruit plants were maintained at The Miracle Fruit Farm in Homestead, FL, USA. Mature plants were maintained in 75.7-L pots with supplemental irrigation under 50% shade. Plants were pruned in February each year to maintain a compact plant structure and fertilized with slow-release fertilizer (18N–2.6P–6.6K) twice per year.
RNA extraction.
RNA was extracted from miracle fruit pulp using the cetyltrimethylammonium bromide method as previously reported, with modifications (Chang et al. 1993). Twenty firm, orange–red berries were selected for extraction from single plants of four different morphotypes, including ‘Sangria’, ‘Vermilion’, ‘Flame’, and ‘Cherry’, that were described previously (Demesyeux et al. 2020). The fruit peel was removed, and the fruit pulp was quickly removed with a dissecting needle and immediately frozen in liquid nitrogen. Exactly 200 mg of pulp from each sample was independently ground to a fine powder using a mortar and pestle, placed in a 5-mL microtube, and 1 mL of extraction buffer [2% cetyltrimethylammonium bromide, 2% polyvinylpyrrolidone (PVP K-30), 100 mM Tris-HCl, 25 mM ethylenediaminetetraacetic acid, 2.0 M NaCl, 0.5 g⋅L−1 spermidine, and 2% 2-mercaptoethanol] was added before incubation at 65 °C for 10 min. Samples were allowed to cool to room temperature, and 1 mL chloroform:isoamyl alcohol (24:1 volume/volume) was added to each tube, vortexed, and centrifuged for 10 min at 8000 gn. The supernatant was aspirated, mixed with 2.0 M lithium chloride, and incubated at −20 °C for 30 min to precipitate RNA. RNA was pelleted by centrifuging at 8000 gn for 30 min at 4 °C. The supernatant was discarded, and the RNA pellet was dissolved in nuclease-free water, treated with DNAse (Qiagen, Hilden, Germany), and column-purified with RNeasy MinElute Cleanup Kit (Qiagen) according to the manufacturer’s instructions.
Transcriptome sequencing.
The four samples were submitted for transcriptome sequencing (RNAseq) with a single replicate of each to obtain a representative pulp transcriptome accounting for possible variations among plant types. These were coded as ‘Sangria’ ACMF2, ‘Vermilion’ ACMF3, ‘Flame’ ACMF5, and ‘Cherry’ ACMF6. For each sample, 2 µg of RNA was submitted to Novogene (Sacramento, CA, USA) for RNAseq on a sequencing machine (HiSeq PE150; Illumina, Inc., San Diego, CA, USA) with 250-bp to 300-bp insert complimentary DNA (cDNA) libraries and up to 20 million reads per sample.
Transcriptome assembly and annotation.
The four RNAseq samples (ACMF2, ACMF3, ACMF5, and ACMF6) were used to create a single, combined transcriptome assembly using Trinity version 2.11.0 (Haas et al. 2013). In brief, reads were processed with Fastq-mcf version 1.05 from the Ea-utils package (Aronesty 2013) to trim adapters and remove low-quality reads (−l 50 −q 30). Each of the datasets was assembled separately with Trinity version 2.11.0 with the default parameters. Supertranscripts were generated from the assemblies with the Trinity_gene_splice_modeler.py script from the Trinity package. A combined assembly was developed with LACE version 1.14.1 (Davidson et al. 2017). Coding DNA sequences and peptides were identified with Transdecoder version 5.5.0 (Haas et al. 2013). Transcript usage was estimated with the dexseq_wrapper.pl after the reads were remapped back to the supertranscripts with STAR version 2.7.5a_2020-06-19 (Dobin et al. 2013). Functional annotation was performed by a sequence homology search using the UniProtKB/SwissProt and UniProtKB/TREMBL datasets (downloaded Nov 2020) (UniProt Consortium 2021) and the Arabidopsis protein sequence set ARAPORT11 with Diamond version 0.9.24 (Buchfink et al. 2021). Both functional annotations were integrated with AHRD version 3.3.3 (Bolger et al. 2017). Gene space completeness on the predicted proteins was evaluated with Benchmarking Universal Single-Copy Orthologue (BUSCO) version 4.1.4 (Simão et al. 2015) with the eudicot_odb10 lineage set.
Variant mining.
Simple sequence repeats (SSRs) in assembled transcripts were mined using GMATA (Wang and Wang 2016) and filtered to include dinucleotide to hexanucleotide repeat motifs with a minimum of five repeats for each SSR motif. To obtain single-nucleotide polymorphism (SNP) markers, adapters were first trimmed from raw sequencing reads with bbduk (Bushnell 2014). Then, the quality reads were mapped to the published miracle fruit genome (China National GeneBank DataBase 2021) using HISAT2 version 2.2 (Kim et al. 2019). Duplicate reads were marked and removed by Picard tools (Broad Institute 2019). Joint variant calling was conducted using GATK version 4.2 with the HaplotypeCaller algorithm in the genomic variant call format (GCVF) mode selecting SNP variants with the -select-type option (Brouard et al. 2019). The final SNPs were filtered using the default options of VariantFiltration (Van der Auwera and O’Connor 2020). Annotation of resulting variants was performed using SNPeff version 5.0e (Cingolani et al. 2012).
Quantitative Polymerase Chain Reaction.
Assembled transcripts were mined for similarity to common housekeeping genes used for the qRT-PCR, such as elongation factor-1 alpha (EF1-α) and actin genes. Assembled transcripts were also screened for sequences similar to the published miraculin gene. Primers for EF1-α, actin, and miraculin were designed with primer3 (Kõressaar et al. 2018) using selected sequences for use during the qRT-PCR analysis. Total RNA was independently extracted in triplicate from miracle fruit tissues as described. The tissues included leaf, root, stem, flowers, and fruit tissues from mature and immature miracle fruit berries (mesocarp, exocarp, seed). The RNA was quantified by a spectrophotometer (NanoDrop 2000; Thermo Fisher Scientific, Waltham, MA, USA) before cDNA preparation using the qPCRBIO cDNA Synthesis Kit (PCR Biosystems, London, UK). The qRT-PCR was conducted in 10-µL reactions containing 0.4 μM of each forward primer and reverse primer, 1 μL of cDNA, and 2x PowerUp SYBR Green Master Mix (Thermo Fisher Scientific). All qRT-PCR experiments were performed using a qRT-PCR thermocycler (Applied Biosystems Quantstudio version 3.0; Thermo Fisher Scientific) using actin primers described in this report as the housekeeping reference gene.
Gene cloning.
The transcriptome was mined for miraculin-like sequences specifically searching for additional 5′ and 3′ untranslated sequences. Primers were designed based on consensus transcripts beyond the start and stop codons of the miraculin gene to simplify the complete cloning of the miraculin gene. Two primers, AC0261 5′-GCTTTCAACTCAGTGCTACGTGGCATT-3′ and AC0262 5′-CCCAGATACTTCGTACACGTCAGAAACAC-3′, were designed at 101 bp upstream and 156 bp downstream of the start and stop codons, respectively. All PCR reactions were performed using VeriFi (PCR Biosystems) polymerase and ‘Flame’ cDNA as the template according to the manufacturer’s instructions. Thermal cycling conditions included an initial denaturation of 4 min at 95 °C, followed by 35 amplification cycles of 30 s at 95 °C, 30 s at 55 °C, and 60 s at 72 °C, with a final 10 min extension at 72 °C. Amplification was confirmed by electrophoresis on a 2% agarose gel. Successfully amplified PCR fragments were purified using EZNA Cycle Pure Kit (Omega Bio-tek Inc., Norcross, GA, USA) and cloned into pJET1.2/blunt plasmid with the CloneJet PCR Cloning Kit (Thermo Fisher Scientific). A microliter of the cloning reaction was added to 50 μL of chemically competent Top10 Escherichia coli, and the mixture was subjected to heat shock at 42 °C for 30 s, followed by immediate cooling on ice for 2 min. Then, cultures were grown at 37 °C in 250 μL Luria-Bertani (LB) medium for 1 h before plating on semi-solid LB with 100 μg⋅mL−1 carbenicillin. The transformed cells were grown overnight at 37 °C; then, the individual colonies were obtained and cultured overnight in 5 mL LB with 100 μg⋅mL−1 carbenicillin. Plasmids were isolated using the EZNA Plasmid DNA Mini Kit I (Omega Bio-tek Inc.) and submitted for Sanger sequencing using pJET1.2 forward and reverse sequencing primers (Genewiz, South Plainfield, NJ, USA).
Results
Sequencing results.
Miracle fruit pulp samples from four plant morphotypes were sequenced to identify abundant transcripts. Overall, the sequencing yield was 79 million reads (Table 1). More than 76 million reads, ranging from 17 to 21 million reads per sample, passed filters and were used for downstream analyses. The guanine and cytosine base contents were similar among samples ranging from 47.1% to 48.2%. There were 36,461 transcripts assembled using the combined RNAseq data (Supplemental Table 1). The individual and combined transcript assemblies were evaluated for completeness and compared with the published miracle fruit genome with BUSCO using eudicots_odb10 lineage (creation date: 10 Sep 2020; number of genomes: 31; number of BUSCOs: 2326). ACMF3 transcript had the most complete BUSCOs (n = 1517), and ACMF2 had the least BUSCOs (n = 1406). ACMF5 and ACMF6 had 1452 and 1499 BUSCOs, respectively. The combined transcriptome assembly contained more complete BUSCOs (n = 1950) than the miracle fruit reference genome (Fig. 1).
Summary statistics of RNA sequencing results of pulp tissues of four miracle fruit morphotypes. Genotypes included ‘Sangria’, ‘Vermilion’, ‘Flame’, and ‘Cherry’. The sample column indicates the coded identity used for bioinformatics analyses, and the raw reads and clean reads columns indicate the total number of raw sequencing reads yielded and reads passing quality filters, respectively.
A comparison of benchmarking universal single-copy orthologue assessment results from individual transcriptome assemblies and a combined assembly of all four transcriptomes and the publicly available miracle fruit reference genome (China National GeneBank DataBase 2021). The evaluation was based on the eudicot_odb10 lineage.
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
Most abundant transcripts.
The most abundant transcripts were analyzed by average counts for the four sequenced samples (Fig. 2, Supplemental Table 1). The most abundant transcript was miraculin (SydulST_CL011578_072032). Other highly abundant genes were putatively annotated as alpha amylase inhibitor (AAI) domain-containing protein (SydulST_CL011578_071958), leucine-rich repeat extensin-like protein 2(SydulST_CL011578_080915), polyubiquitin (SydulST_CL011578_072822), UDP-glucuronate 4-epimerase (SydulST_CL011578_072903), beta-galactosidase (SydulST_CL011578_072145), polyubiquitin (SydulST_CL011578_073485, SydulST_CL011578_139377), thaumatin-like protein (SydulST_CL011578_072810), glutamate decarboxylase (SydulST_CL011578_072109), and calreticulin (SydulST_CL011578_071638). Eukaryotic translation initiation factor-1 alpha (SydulST_CL011578_064409) and actin (SydulST_CL011578_076296) were also highly abundant and useful for designing primers for qRT-PCR housekeeping genes.
Ten genes with the highest abundance of transcripts from miracle fruit pulp tissues. Transcript abundance for eukaryotic translation initiation factor-1α and actin (used for real-time quantitative reverse-transcription polymerase chain reaction housekeeping genes) are also shown. Bars on the columns indicate errors across averages for the four genotypes. Lowercase letters above the bars indicate significant difference at P < 0.05 using Tukey’s test.
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
Identification of SSR variants.
The SSR variants are still used for diversity and seedling analysis of some species. In contrast to SSRs mined from a genome, SSRs from a transcriptome are interesting because they are expressed in messenger RNAs. A total of 15,162 SSR markers were identified within the miracle fruit consensus transcriptome. The SSR motif sizes ranged from 2 to 6 bp, with repeat numbers from 5 to 20 (Fig. 3). In general, the number of SSRs identified, motif length, and number of repeats decreased as complexity increased with a rapid decline for 4-mers and above.
The number and type of short sequence repeat (SSR) markers identified in the coding regions of the consensus miracle fruit transcriptome. Plot columns are color-coded according to SSR motif repetitions of each SSRs type.
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
Comparison of morphotype transcriptomes.
Comparing the transcriptomes for the individual morphotypes can identify transcript differences that could have biological implications. A comparison of transcripts revealed unique transcripts in each of the four miracle fruit morphotypes (Fig. 4). The most unique transcripts were observed in ACFM3 (n = 1122), and the least were observed in ACFM6 (n = 850). ACFM2 and ACFM 5 had 1029 and 876 unique transcripts, respectively. All four morphotypes shared 1912 transcripts.
Venn diagram indicating the number of unique and shared transcripts among four miracle fruit morphotypes. Samples are coded as ‘Sangria’ (ACMF2), ‘Vermilion’ (ACMF3), ‘Flame’ (ACMF5), and ‘Cherry’ (ACMF6).
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
Identification of SNP variants.
The SNP variants are necessary for the diversity analysis, marker trait associations, and other related genetic analyses. For SNP variant discovery, RNAseq reads were mapped to the published miracle fruit genome. The total number of genetic variants after quality filtering among the four genotypes was 13,630, with 8093 exhibiting polymorphisms (Supplemental Tables 2–5). Polymorphic variants included 6661 insertions, 1135 SNPs, and 297 deletions. Most of the variants were located within introns (n = 3283), with 2146 variants occurring upstream of the predicted protein coding sequence and 2643 occurring downstream of the predicted protein coding sequence. A total of 426 variants occurring in exons were characterized as silent (31.9%), missense (65.4%), or nonsense (2.7%) mutations. For SNPs in exons, 8.5% were predicted to have a high impact, 13.1% were predicted to have a low impact, 1.1% were predicted to have low impact, and the majority (77.3%) had a modifier impact. The variant rate across the chromosomes ranged from one SNP per 51.8 to 83.6 bp for all chromosomes except chromosome 11, with a rate of one SNP per 124.1 (Supplemental Table 2).
Qualitative transcript comparison and qRT-PCR of target genes.
The qRT-PCR analysis confirmed miraculin transcript abundance in the exocarp (skin) and mesocarp (pulp) of mature and immature miracle fruit berries (Fig. 5). There was minimal transcript abundance in flowers or seed, and minimal transcript abundance in roots, leaves, and stem tissues (Fig. 5).
Real-time quantitative reverse-transcription polymerase chain reaction transcript abundance profiles for the miraculin gene target in different tissues of miracle fruit plant. Values represent averages from three biological replicates with the SE shown. Lowercase letters indicate significant difference at P < 0.05 by Tukey’s test.
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
Miraculin gene cloning and variants.
Filtered RNAseq reads were subjected to a homology search using Basic Local Alignment Search Tool (BLAST) of the National Center for Biotechnology Information (Bethesda, MD, USA) (Sayers et al. 2022) to retrieve miraculin-like sequences from each of the accessions. Then, the sequences were compared with published miraculin sequences (Genbank ID D38598 and AB512278.1) to analyze any potential sequence diversity. There were 1.01, 1.02, 0.74, and 0.79 million reads matching the miraculin gene for ‘Sangria’ ACMF2, ‘Vermilion’ ACMF3, ‘Flame’ ACMF5, and ‘Cherry’ ACMF6, respectively. A single SNP (T/C) at bp 102 (from the adenine in the start codon) was identified and showed variability among the four morphotypes (Fig. 6). This is a silent mutation with both alleles coding for asparagine. ‘Sangria’ ACMF2 had the reference T allele at 0.3%, whereas the other accessions were 36.7%, 44.6%, and 46.8% for the reference allele for ‘Vermilion’ ACMF3, ‘Flame’ ACMF5, and ‘Cherry’ ACMF6, respectively. No other SNPs or other variants were common among the accessions within the miraculin-like sequences.
Miraculin allelic variants sequenced as part of this study in comparison with published sequences. Sequences from this study are labeled as Miracle_Fruit_C_allele or Miracle_Fruit_T_allele. Reference sequences include D38598 and AB512278 from the National Center for Biotechnology Information (Bethesda, MD, USA) database. The black box outlines the single single-nucleotide polymorphism identified in the transcriptomes and confirmed using Sanger sequencing. Predicted protein translations are shown for both allelic variants.
Citation: Journal of the American Society for Horticultural Science 148, 5; 10.21273/JASHS05312-23
There is limited, publicly available sequence information for the upstream and downstream regions of the miraculin gene. The transcriptome assembly enabled the expansion of the 5′ and 3′ untranslated regions of the miraculin gene. Using this information, primers AC0261 and AC0262 were designed at 101 bp upstream and 156 bp downstream of the start and stop codons, respectively. The resulting PCR product was 920 bp, which was suitable for sequencing of the full coding region of the miraculin gene without primers masking any potential SNPs in the gene itself. Sanger sequencing of the miraculin gene confirmed the presence of both allelic variants at bp 102.
Discussion
Miracle fruit has great potential to decrease sugar consumption based on the effects of miraculin, the desirable active ingredient of miracle fruit. The fruit from this species are attractive and have the potential for high economic returns for growers. However, the species is limited by a lack of awareness, rapid postharvest fruit decline, and the sensitivity of the miraculin protein to denaturing during food product manufacturing. Additional studies and investment are also needed to qualify miracle fruit to have a generally recognized as safe status according to the United States Food and Drug Administration. In the future, the availability of genomics datasets for miracle fruit will enable researchers worldwide to uncover plant biological processes for this “orphan” crop. These resources illuminate possibilities, generate hypotheses, and facilitate additional molecular and biochemical studies as a first step toward improved genetics of miracle fruit.
The transcriptomes in this study confirmed that the miraculin gene is the most abundant transcript found in fresh fruit. Although the role of miraculin in an ecological or evolutionary context is still not fully understood, the abundance of the desired active ingredient is favorable compared with that of some natural, noncaloric metabolites that are only present at low levels. These results seem to be consistent across multiple research groups and plant morphotypes (Demesyeux et al. 2020; Yang et al. 2022). This is a favorable situation for those seeking to grow miracle fruit for the active ingredient, and future work could identify the correlation between miraculin transcript and protein abundance perhaps among accessions in the center of diversity.
In addition to miraculin, other highly abundant transcripts could also be interesting to the biological processes of the miracle fruit and support hypothesis generation. For example, calreticulin was an abundant transcript in miracle fruit pulp and plays a role in calcium ion sequestration (Jia et al. 2009; Persson et al. 2001). Miracle fruit is reportedly high in calcium (Awotedu and Ogunbamowo 2019). The concentration of miracle fruit Ca2+ and calreticulin could be correlated, but future work is needed to resolve this hypothesis. The other abundant transcripts are the AAI domains. Plants produce AAI domains to inhibit starch metabolism by pests and pathogens (Franco et al. 2002). Additionally, AAI domains of plant products are used to inhibit glucose metabolism and reduce blood sugar levels of diabetic patients (Khadayat et al. 2020; Sudha et al. 2011). In fact, similar antidiabetic properties have been observed in miracle fruit pulp extracts through inhibition α-amylase and α-glucosidase enzymes (Fazilah et al. 2020).
Advancements in miracle fruit genomics and interest in plant breeding of this species are encouraging. Much of this work is being conducted in the center of diversity in Africa. The development of SNP markers will benefit many studies of this species. The unique SNP sets identified during this study could be used to genotype miracle fruit collections previously characterized for fruit quality and yield (Demesyeux et al. 2020). This would help resolve previous challenges arising from unknown genetics that complicate the grouping of morphotypes for data analysis. The identified SNPs could also be used to create low-cost genotyping assays for diversity collections in Africa. Additionally, SNPs located in exons could have biological implications that are worth pursuing in the future.
This study also designed and validated PCR primers for cloning the entire coding sequence of the miraculin gene. Although no new variants were discovered in the four accessions tested during this study, the primers reported could test for additional miraculin gene variants in a diversity collection. Any potential variants identified could have interesting impacts on the sensory level. Variants in the miraculin coding sequence could be useful tools to further advance breeding work in this species.
Miracle fruit is an intriguing species that has benefited from researcher interest. The genomics work performed during this study will benefit future studies aimed at the genetic improvement of this species. Future economic opportunities for miracle fruit could depend on genetic advancements of this species.
References Cited
Achigan-Dako EG, Tchokponhoué DA, N’Danikou S, Gebauer J, Vodouhè RS. 2015. Current knowledge and breeding perspectives for the miracle plant Synsepalum dulcificum (Schum. et Thonn.) Daniell. Genet Resources Crop Evol. 62(3):465–476. https://doi.org/10.1007/s10722-015-0225-7.
Aronesty E. 2013. Comparison of sequencing utility programs. Open Bioinform J. 7(1):1–8. https://doi.org/10.2174/1875036201307010001.
Awotedu OL, Ogunbamowo PO. 2019. Nutritional, anti-nutritional and phytochemical profile of the leaves and fruits of Synsepalum dulcificum (Schumach.& Thonn.) Daniell. Am J Biol Chem. 7(3):53–59.
Bolger ME, Arsova B, Usadel B. 2017. Plant genome and transcriptome annotations: From misconceptions to simple solutions. Brief Bioinform. 19(3):437–449. https://doi.org/10.1093/bib/bbw135.
Broad Institute. 2019. Picard tools. https://broadinstitute.github.io/picard/. [accessed 25 May 2023].
Brouard J-S, Schenkel F, Marete A, Bissonnette N. 2019. The GATK joint genotyping workflow is appropriate for calling variants in RNA-seq experiments. J Anim Sci Biotechnol. 10(1):44. https://doi.org/10.1186/s40104-019-0359-0.
Buchfink B, Reuter K, Drost H-G. 2021. Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat Methods. 18(4):366–368. https://doi.org/10.1038/s41592-021-01101-x.
Bushnell B. 2014. BBMap. https://sourceforge.net/projects/bbmap/. [accessed 13 Jul 2023].
Chang S, Puryear J, Cairney J. 1993. A simple and efficient method for isolating RNA from pine trees. Plant Mol Biol Rpt. 11(2):113–116. https://doi.org/10.1007/BF02670468.
China National GeneBank DataBase. 2021. Miracle fruit genome. doi:10.26036/CNP0002330. https://db.cngb.org/search/project/CNP0002330/. [accessed 19 May 2023].
Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, Land SJ, Lu X, Ruden DM. 2012. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff. Fly (Austin). 6(2):80–92. https://doi.org/10.4161/fly.19695.
Davidson NM, Hawkins ADK, Oshlack A. 2017. SuperTranscripts: A data driven reference for analysis and visualisation of transcriptomes. Genome Biol. 18(1):148. https://doi.org/10.1186/s13059-017-1284-1.
Demesyeux L, Brym M, Valdes D, Collazo C, Chambers AH. 2020. Yield and miraculin content of nine miracle fruit (Synsepalum dulcificum) morphotypes. Euphytica. 216(11):181. https://doi.org/10.1007/s10681-020-02710-x.
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. 2013. STAR: Ultrafast universal RNA-seq aligner. Bioinformatics. 29(1):15–21. https://doi.org/10.1093/bioinformatics/bts635.
Fazilah NF, Ariff AB, Khayat ME, Halim M. 2020. Anti-diabetic properties of Synsepalum dulcificum and its potential inclusion in functional yogurt. IOP Conf Ser Mater Sci Eng. 716(1):012004. doi:10.1088/1757-899X/716/1/012004.
Franco OL, Rigden DJ, Melo FR, Grossi-de-Sá MF. 2002. Plant α-amylase inhibitors and their interaction with insect α-amylases. Eur J Biochem. 269(2):397–412. https://doi.org/10.1046/j.0014-2956.2001.02656.x.
Gibbs BF, Alli I, Mulligan C. 1996. Sweet and taste-modifying proteins: A review. Nutr Res. 16(9):1619–1630. https://doi.org/10.1016/0271-5317(96)00175-3.
Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, Couger MB, Eccles D, Li B, Lieber M, MacManes MD, Ott M, Orvis J, Pochet N, Strozzi F, Weeks N, Westerman R, William T, Dewey CN, Henschel R, LeDuc RD, Friedman N, Regev A. 2013. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 8(8):1494–1512. https://doi.org/10.1038/nprot.2013.084.
Inglett GE, Dowling B, Albrecht JJ, Hoglan FA. 1965. Taste modifiers, taste-modifying properties of miracle fruit (Synsepalum dulcificum). J Agr Food Chem. 13(3):284–287. https://doi.org/10.1021/jf60139a026.
Jia X-Y, He L-H, Jing R-L, Li R-Z. 2009. Calreticulin: Conserved protein and diverse functions in plants. Physiol Plant. 136(2):127–138. https://doi.org/10.1111/j.1399-3054.2009.01223.x.
Kant R. 2005. Sweet proteins – Potential replacement for artificial low calorie sweeteners. Nutr J. 4(1):5. https://doi.org/10.1186/1475-2891-4-5.
Khadayat K, Marasini BP, Gautam H, Ghaju S, Parajuli N. 2020. Evaluation of the alpha-amylase inhibitory activity of Nepalese medicinal plants used in the treatment of diabetes mellitus. Clin Phytoscience. 6(1):34. https://doi.org/10.1186/s40816-020-00179-8.
Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. 2019. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 37(8):907–915. https://doi.org/10.1038/s41587-019-0201-4.
Kõressaar T, Lepamets M, Kaplinski L, Raime K, Andreson R, Remm M. 2018. Primer3_masker: Integrating masking of template sequence with primer design software. Bioinformatics. 34(11):1937–1938. https://doi.org/10.1093/bioinformatics/bty036.
Kurihara K, Beidler LM. 1968. Taste-modifying protein from miracle fruit. Science. 161(3847):1241–1243. https://doi.org/10.1126/science.161.3847.1241.
Malik VS, Willett WC, Hu FB. 2013. Global obesity: Trends, risk factors and policy implications. Nat Rev Endocrinol. 9(1):13–27. https://doi.org/10.1038/nrendo.2012.199.
Niu Y, Ni S, Liu J. 2020. Complete chloroplast genome of Synsepalum dulcificum D.: A magical plant that modifies sour flavors to sweet. Mitochondrial DNA B Resour. 5(3):3052–3053. https://doi.org/10.1080/23802359.2020.1798299.
Persson S, Wyatt SE, Love J, Thompson WF, Robertson D, Boss WF. 2001. The Ca2+ status of the endoplasmic reticulum is altered by induction of calreticulin expression in transgenic plants. Plant Physiol. 126(3):1092–1104. https://doi.org/10.1104/pp.126.3.1092.
Rodrigues JF, Andrade R da S, Bastos SC, Coelho SB, Pinheiro ACM. 2016. Miracle fruit: An alternative sugar substitute in sour beverages. Appetite. 107:645–653. https://doi.org/10.1016/j.appet.2016.09.014.
Sayers EW, Bolton EE, Brister JR, Canese K, Chan J, Comeau DC, Connor R, Funk K, Kelly C, Kim S, Madej T, Marchler-Bauer A, Lanczycki C, Lathrop S, Lu Z, Thibaud-Nissen F, Murphy T, Phan L, Skripchenko Y, Tse T, Wang J, Williams R, Trawick BW, Pruitt KD, Sherry ST. 2022. Database resources of the National Center for Biotechnology Information. Nucleic Acids Res. 50(D1):D20–D26. https://doi.org/10.1093/nar/gkab1112.
Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. 2015. BUSCO: Assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. 31(19):3210–3212. https://doi.org/10.1093/bioinformatics/btv351.
Smith KB, Smith MS. 2016. Obesity statistics. Prim Care Clin Off Pract. 43(1):121–135. https://doi.org/10.1016/j.pop.2015.10.001.
Sudha P, Zinjarde SS, Bhargava SY, Kumar AR. 2011. Potent α-amylase inhibitory activity of Indian Ayurvedic medicinal plants. BMC Complement Altern Med. 11(1):5. https://doi.org/10.1186/1472-6882-11-5.
Temussi PA. 2006. Natural sweet macromolecules: How sweet proteins work. Cell Mol Life Sci. 63(16):1876–1888. https://doi.org/10.1007/s00018-006-6077-8.
Theerasilp S, Hitotsuya H, Nakajo S, Nakaya K, Nakamura Y, Kurihara Y. 1989. Complete amino acid sequence and structure characterization of the taste-modifying protein, miraculin. J Biol Chem. 264(12):6655–6659.
UniProt Consortium. 2021. UniProt: The universal protein knowledgebase in 2021. Nucleic Acids Res. 49(D1):D480–D489. https://doi.org/10.1093/nar/gkaa1100.
Van der Auwera GA, O’Connor BD. 2020. Genomics in the cloud: Using Docker, GATK, and WDL in Terra. O’Reilly Media, Inc., Sebastopol, CA, USA.
Wang X, Wang L. 2016. GMATA: An integrated software package for genome-scale SSR mining, marker development and viewing. Front Plant Sci. 7. https://doi.org/10.3389/fpls.2016.01350.
Wilken MK, Satiroff BA. 2012. Pilot study of “miracle fruit” to improve food palatability for patients receiving chemotherapy. Clin J Oncol Nurs. 16(5):E173–E177. https://doi.org/10.1188/12.CJON.E173-E177.
Xingwei C, Abdullah TL, Taheri S, Abdullah NAP, Hassan SA. 2016. Flower ontogenesis and fruit development of Synsepalum dulcificum. HortScience. 51(6):697–702. https://doi.org/10.21273/HORTSCI.51.6.697.
Yang Z, Liu Z, Xu H, Chen Y, Du P, Li P, Lai W, Hu H, Luo J, Ding Y. 2022. The Chromosome-level genome of miracle fruit (Synsepalum dulcificum) provides new insights into the evolution and function of miraculin. Front Plant Sci. 12. https://doi.org/10.3389/fpls.2021.804662. https://www.frontiersin.org/articles/10.3389/fpls.2021.804662/full.